[PATCH 4/8] score: Improve _CORE_message_queue_Initialize()

Sebastian Huber sebastian.huber at embedded-brains.de
Fri Sep 25 04:32:55 UTC 2020


On 24/09/2020 18:39, Gedare Bloom wrote:

>> +  status = _CORE_message_queue_Initialize(
>> +    &the_mq->Message_queue,
>> +    CORE_MESSAGE_QUEUE_DISCIPLINES_FIFO,
>> +    attr->mq_maxmsg,
>> +    attr->mq_msgsize
>> +  );
>> +
>> +  if ( status != STATUS_SUCCESSFUL ) {
>>       _POSIX_Message_queue_Free( the_mq );
>>       _Workspace_Free( name );
>>       rtems_set_errno_and_return_value( ENOSPC, MQ_OPEN_FAILED );
> translate the status in case we end up using something besides ENOSPC?
>
I checked the standard and there is no explicit error condition listed 
for sizes which lead to an integer overflow. If you do the calculations 
with real integers, then you can say, that a 4GB address space system 
has insufficient space for the creation of the new message queue needing 
5GB.


More information about the devel mailing list